August 2006 - U-Boot

Re: [U-Boot-Users] TFTP timeout
by Olivier Vernet 02 Aug '06

02 Aug '06

> >Then the next thing you should do is updating to top of tree. > >You should always use *current* code for new projects. Hi everyone, I'm working with Philippe on the same board, we've just download top-of-git uboot and we have the same problem with TFTP. No more idea? Best regards Olivier

2 1

[U-Boot-Users] 440SP DDR2 Issue
by Jeff Stevens 02 Aug '06

02 Aug '06

I have a custom board I am trying to bring up that is a pretty close architecture to the Luan2 board. I have an unregistered 2GB DDR2 module in the DIMM slot, and it works fine using the BDI2000. However, using the same DDR2 configuration in U-Boot doesn't work. It boots up until the point it transfers over to memory. I have enabled the memory test and it fails at the first address. I have verified all of the memory configuration registers, and all seem to be set correctly. I know there is something in the configuration I am missing, but I can't seem to figure it out. Please help. I would greatly appreciate it. Thanks, Jeff Stevens __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

3 2

[U-Boot-Users] NAND flash support
by mitsy 01 Aug '06

01 Aug '06

Hi, I've been trying to add NAND flash support to my u-boot but having issues. I'd really appreciate if anyone could give me any suggestions/comments on what I'm doing wrong. The board I have is based on AMD au1550, and it has one NOR flash and one NAND flash. NAND flash is configured in lowlevel_init.S correctly (I believe so since it works from Linux kernel.) However, when I try to send any command to NAND flash, it just hangs there. Here are the command macros I have: #define NAND_WAIT_READY(nand) while (!(*(volatile u8 *)((ulong)MEM_STSTAT) & 0x01)) #define WRITE_NAND_COMMAND(d, adr) do { printf("weee!"); *(volatile u8 *)((ulong)adr+MEM_STNAND_CMD) = d; __asm__ volatile ("sync"); } while(0) #define WRITE_NAND_ADDRESS(d, adr) do {*(volatile u8 *)((ulong)adr+MEM_STNAND_ADDR) = d; __asm__ volatile ("sync");} while(0) #define WRITE_NAND(d, adr) do {*(volatile u8 *)((ulong)adr+MEM_STNAND_DATA) = d; __asm__ volatile ("sync");} while(0) #define READ_NAND(adr) (*(volatile u8 *)((ulong)adr+MEM_STNAND_DATA)) #define NAND_DISABLE_CE(nand) do { *(volatile u8 *)(MEM_STNDCTL) = 0; __asm__ volatile ("sync");} while(0) #define NAND_ENABLE_CE(nand) do { *(volatile u8 *)(MEM_STNDCTL) = (1<<5); __asm__ volatile ("sync");} while(0) These are empty since the processor handles them. #define NAND_CTL_CLRALE #define NAND_CTL_SETALE #define NAND_CTL_CLRCLE #define NAND_CTL_SETCLE Thank you for your help in advance! Mitsy

1 0

[U-Boot-Users] Help on initrd Ramdisk filesystem!
by Kylong Mu 01 Aug '06

01 Aug '06

Hi, Team I build succesfull with uboot1.4 and 2.6.18 kernel on my at91rm9200ek board, and mount through the default ELDK NFS filesystem is ok. And I want build an initrd Ramdisk filesystem, but I find it's hard to find an example can run under uboot1.4 and 2.6.18 kernel, if anybody did successful same with my seroundings, please mail your filesystem image to me, or thanks for any advice! Cordially, KylongMu

2 1

Re: [U-Boot-Users] booting elf image?
by Wolfgang Denk 01 Aug '06

01 Aug '06

[Please keep the list on cc:] In message <8bf247760608010750n4da0cde9u376a2b521ed05187(a)mail.gmail.com> you wrote: > > That is the reason i have built executable which is statically linked. > > isnt this enough?. No, of course not. As I wrote in my first reply, you need at least a basic understanding of the operating system, of the system services provided by an OS, of the libraries that are based on these services, and of the run-time environments for C programs. In your code you use the printf() function - guess what it does? It will boil down to using some system calls - like write(). This is a service provided by your OS. Also it relies on concepts like standard I/O streams (namely, stdout), which in turn relies on the standard file descriptors 0, 1, and 2 being set up by your OS. The whole concept of a "process" and it's environment (virtual address space, open file descriptors, etc.) is highly OS dependent. Do you think another OS (say, Windoze) would provide exactly teh same (binary compatible!) system call interface? Or process environment? And do you really think a simple boot loader like U-Boot would provide such services? Well, let me tell you: it ain't so. Your a.out program will only run in a Linux environment (or in one which is compatible - for example, it is possible to run Linux binaries under FreeBSD). Best regards, Wolfgang Denk -- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd(a)denx.de Nothing ever becomes real until it is experienced. - John Keats

2 1

[U-Boot-Users] booting elf image?
by Ram 01 Aug '06

01 Aug '06

i, i am using omap 5912osk. (ARM architecture). I have a C program: int main ( ) { printf ("Hello World\n") ; } im compiling the c program with static linking. [sriram@sriram sriram]# file a.out a.out: ELF 32-bit LSB executable, ARM, version 1 (ARM), for GNU/Linux 2.4.3, statically linked, not stripped I download the elf executable using tftpboot onto my target . when i do bootelf u-boot hangs? Am i missing something? Regards, sriram

3 3

[U-Boot-Users] Uboot version querry
by Ajay Jain 01 Aug '06

01 Aug '06

Hi, We are porting uboot to an ARM based proprietary board (MM6+). I was looking at the version available on the ftp site. It has the latest version "u-boot-2006-06-30-2020.tar.bz2", whereas the latest version that sourceforge talks about is "U-Boot 1.1.4". Which is the recommended stable version, to begin the port. I could not find this information anywhere on the uboot official site. Another querry, other than the "README", is there any other document availabe that talks about porting uboot on new ARM board? Thanks, Ajay

2 1

[U-Boot-Users] UBOOT - Arm Floating Point
by Carlos Coelho 01 Aug '06

01 Aug '06

Hi, I am doing the following commands: make mrproper OK make at91rm9200dk_config OK make u-boot and I=B4ve got the following error message: ... -Map u-boot.map -o u-boot arm-linux-ld: ERROR: = /usr/local/arm/3.4.1/lib/gcc/arm-linux/3.4.1/libgcc.a(_divsi3.oS) uses = hardware FP, whereas u-boot uses software FP The "cpu/arm920t/config.mk" file has the "-msoft-float" option set, but = the arm-linux-gcc seems to be compiling everything for a hardware FP. Does someone had this problem before?=20 Regards, Carlos

5 6

[U-Boot-Users] [PATCH] for FLASH sector info reverse order & NS16550 pre-setting PSR
by Roger Huang 01 Aug '06

01 Aug '06

HI: I use a 8MB TOP FLASH, with CFI driver can not handle this type, becasue SECT info order is for BOTTOM FLASH, so I modefied for it. Next version can detect bootloc :-) Enjoy Roger Huang diff -rc u-boot/drivers/cfi_flash.c u-boot.cameo/drivers/cfi_flash.c *** u-boot/drivers/cfi_flash.c 2006-07-31 14:25:47.000000000 +0800 --- u-boot.cameo/drivers/cfi_flash.c 2006-07-31 12:39:55.000000000 +0800 *************** *** 1173,1179 **** } tmp = flash_read_long (info, 0, FLASH_OFFSET_ERASE_REGIONS + ! i * 4); erase_region_size = (tmp & 0xffff) ? ((tmp & 0xffff) * 256) : 128; tmp >>= 16; --- 1173,1184 ---- } tmp = flash_read_long (info, 0, FLASH_OFFSET_ERASE_REGIONS + ! #ifdef CFG_FLASH_SECT_INFO_REVS ! (num_erase_regions - i - 1) * ! #else ! i * ! #endif ! 4); erase_region_size = (tmp & 0xffff) ? ((tmp & 0xffff) * 256) : 128; tmp >>= 16; diff -rc u-boot/drivers/ns16550.c u-boot.cameo/drivers/ns16550.c *** u-boot/drivers/ns16550.c 2006-07-31 14:25:47.000000000 +0800 --- u-boot.cameo/drivers/ns16550.c 2006-07-27 13:08:00.000000000 +0800 *************** *** 23,28 **** --- 23,31 ---- com_port->lcr = LCR_BKSE | LCRVAL; com_port->dll = baud_divisor & 0xff; com_port->dlm = (baud_divisor >> 8) & 0xff; + #ifdef CFG_NS16550A_PSR + com_port->fcr = CFG_NS16550A_PSR; + #endif com_port->lcr = LCRVAL; com_port->mcr = MCRVAL; com_port->fcr = FCRVAL; diff -rc u-boot/README u-boot.cameo/README *** u-boot/README 2006-07-31 14:25:45.000000000 +0800 --- u-boot.cameo/README 2006-07-31 12:39:06.000000000 +0800 *************** *** 1818,1823 **** --- 1818,1826 ---- is useful, if some of the configured banks are only optionally available. + - CFG_FLASH_SECT_INFO_REVS + Reverse sector infomation order for broken CFI top devices + - CFG_RX_ETH_BUFFER: Defines the number of ethernet receive buffers. On some ethernet controllers it is recommended to set this value

2 1